{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CZPRYVZN5B5OH6WKK5DEUBDGVW","short_pith_number":"pith:CZPRYVZN","schema_version":"1.0","canonical_sha256":"165f1c572de87ae3faca57464a0466adadfff3df0902c93ca52b9c37506fe92f","source":{"kind":"arxiv","id":"2507.04049","version":4},"attestation_state":"computed","paper":{"title":"DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.CV","authors_text":"Bencheng Liao, Caiyan Jia, Hongyu Pan, Lei Yang, Lin Liu, Mingzhe Guo, Shaoqing Xu, Yadan Luo, Yongchang Zhang, Ziying Song","submitted_at":"2025-07-05T14:19:19Z","abstract_excerpt":"Most end-to-end autonomous driving methods rely on imitation learning from single expert demonstrations, often leading to conservative and homogeneous behaviors that limit generalization in complex real-world scenarios. In this work, we propose DIVER, an end-to-end driving framework that integrates reinforcement learning with diffusion-based generation to produce diverse and feasible trajectories. At the core of DIVER lies a reinforced diffusion-based generation mechanism. First, the model conditions on map elements and surrounding agents to generate multiple reference trajectories from a sing"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2507.04049","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-05T14:19:19Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"e88eae88c0f4151c2893fe9c023e51bacbd1963c7ce583ee5945abdfe333e1ec","abstract_canon_sha256":"2161ede9e719997bb951a38a3a33e9bb3467968b4bf523a1139f671aefa437d0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:04.882357Z","signature_b64":"Enpea1p9Z5G34uK7fT7OjlR9YCcRdVdR3f0aamN102fpxfmNf5aiqgge/sh+l4M5HduhmqUF+6HouVxJ+vqeBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"165f1c572de87ae3faca57464a0466adadfff3df0902c93ca52b9c37506fe92f","last_reissued_at":"2026-05-17T23:39:04.881626Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:04.881626Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.CV","authors_text":"Bencheng Liao, Caiyan Jia, Hongyu Pan, Lei Yang, Lin Liu, Mingzhe Guo, Shaoqing Xu, Yadan Luo, Yongchang Zhang, Ziying Song","submitted_at":"2025-07-05T14:19:19Z","abstract_excerpt":"Most end-to-end autonomous driving methods rely on imitation learning from single expert demonstrations, often leading to conservative and homogeneous behaviors that limit generalization in complex real-world scenarios. In this work, we propose DIVER, an end-to-end driving framework that integrates reinforcement learning with diffusion-based generation to produce diverse and feasible trajectories. At the core of DIVER lies a reinforced diffusion-based generation mechanism. First, the model conditions on map elements and surrounding agents to generate multiple reference trajectories from a sing"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.04049","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2507.04049","created_at":"2026-05-17T23:39:04.881757+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.04049v4","created_at":"2026-05-17T23:39:04.881757+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.04049","created_at":"2026-05-17T23:39:04.881757+00:00"},{"alias_kind":"pith_short_12","alias_value":"CZPRYVZN5B5O","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"CZPRYVZN5B5OH6WK","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"CZPRYVZN","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":5,"sample":[{"citing_arxiv_id":"2605.13646","citing_title":"Causality-Aware End-to-End Autonomous Driving via Ego-Centric Joint Scene Modeling","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15120","citing_title":"CLOVER: Closed-Loop Value Estimation and Ranking for End-to-End Autonomous Driving Planning","ref_index":31,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13646","citing_title":"Causality-Aware End-to-End Autonomous Driving via Ego-Centric Joint Scene Modeling","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09701","citing_title":"DriveFuture: Future-Aware Latent World Models for Autonomous Driving","ref_index":49,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17841","citing_title":"Driving risk emerges from the required two-dimensional joint evasive acceleration","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CZPRYVZN5B5OH6WKK5DEUBDGVW","json":"https://pith.science/pith/CZPRYVZN5B5OH6WKK5DEUBDGVW.json","graph_json":"https://pith.science/api/pith-number/CZPRYVZN5B5OH6WKK5DEUBDGVW/graph.json","events_json":"https://pith.science/api/pith-number/CZPRYVZN5B5OH6WKK5DEUBDGVW/events.json","paper":"https://pith.science/paper/CZPRYVZN"},"agent_actions":{"view_html":"https://pith.science/pith/CZPRYVZN5B5OH6WKK5DEUBDGVW","download_json":"https://pith.science/pith/CZPRYVZN5B5OH6WKK5DEUBDGVW.json","view_paper":"https://pith.science/paper/CZPRYVZN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.04049&json=true","fetch_graph":"https://pith.science/api/pith-number/CZPRYVZN5B5OH6WKK5DEUBDGVW/graph.json","fetch_events":"https://pith.science/api/pith-number/CZPRYVZN5B5OH6WKK5DEUBDGVW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CZPRYVZN5B5OH6WKK5DEUBDGVW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CZPRYVZN5B5OH6WKK5DEUBDGVW/action/storage_attestation","attest_author":"https://pith.science/pith/CZPRYVZN5B5OH6WKK5DEUBDGVW/action/author_attestation","sign_citation":"https://pith.science/pith/CZPRYVZN5B5OH6WKK5DEUBDGVW/action/citation_signature","submit_replication":"https://pith.science/pith/CZPRYVZN5B5OH6WKK5DEUBDGVW/action/replication_record"}},"created_at":"2026-05-17T23:39:04.881757+00:00","updated_at":"2026-05-17T23:39:04.881757+00:00"}